HPE 3PAR StoreServ Reviews

We use the solution for multiple business applications, with SharePoint, our internal management system and Exchange 2016. We are customers of HPE 3PAR and I'm a system engineer.

I think the main thing of value to me is the intelligence around the solution and how data is taken care of on the storage side of things, with your LANs and your virtual volumes and how it sort of manages everything else.

The scalability could be improved plus management. Currently, you can only identify what's connected to it through the system names and the like and it would be helpful to have an overview of everything that comes to it. It's easy to upsize but you can't downsize.

We've been using this solution for about seven years. 

This is a stable solution. 

The product used to be very scalable but its end of life will be in about a year so there's no point to upgrade or enhance it. 

The support is quite good and we'll continue to use this product.

The initial setup is complex, it requires you to do your volumes, fibre channel volume from the switch and the same connection from the storage. You also need to make sure you've created warning profiles. I don't recall how long the primary setup took, but the secondary one took us four to six hours to get it going. We used a consultant and we have 1,200 users.
We have about six people involved in maintenance, they are admins.

The product is at end of life so although we'll continue to use it, I wouldn't recommend it. 

I rate this solution an eight out of 10. 

The primary use case of this solution is for clusters, database work (Oracle and MySQL), and storage.

An area of improvement for this solution is an increase in the bandwidth as well as an upgrade of the storage functionality and capabilities. The storage needs to be expandable for future-proofing.

An additional feature I would like to see in the next release is for the management interface not to depend on Java.

I have been using the solution for three and a half years.

The solution is easy to use and very stable.

The solution is scalable, but we required an extension.

The technical support is very good.

Previously, I used EMC VNX. We switched because we couldn't monitor anything unless we uploaded a data file to EMC and they send us back an analysis.

The installation of the solution required three people and took three days, but for the storage-side, it took eight hours to get running.

We use this solution for mixed, unpredicted workloads.

This solution has allowed for massive performance acceleration of all workloads and massively increased availability (with peer persistence/transparent failover feature).

The most valuable features are:

• Peer Persistence, where storage failover is transparent to servers and there is no complex cluster-software integration.
  
• Thin Provisioning translates to reduced efforts in management on the server-side, where calculated growth of application data and LUNs can be provisioned on the first install.

The Unified Multiprotocol Access to the storage array needs to be improved.

I found it easy to deploy and simple to configure the storage.

We have had no problems so far.

One of the points that moved us towards 3PAR was we wanted to escalate it in a very high way.

We have not used tech support often, but when we have, they have been clear with us.

We have already some IBM storage, but we are not happy with it. After checking out other models, we decided to work with 3PAR. Not only because we are familiar with HPE, but also the features combined well with our services.

The initial setup was easy. However, we get stuck on preconditions. We were not aware of some of the preconditions. For example, the need for a different, separate IP network for replication. We had to rethink how we were going to implement it.

It has provided a robust technology which provides us with fast access to the data. This is one of the areas where it has improved our company.

3PAR is very fast and robust.

I have been looking into additional functionalities of 3PAR but at this moment we are very satisfied with the basic functionality.

It's stable. I can't recall any problem with the 3PAR technology.

It's scalable. We are creating a plan, with our HPE partner in Brazil to duplicate our 3PAR environment because for us it is very important. It's the base of our data center.

Support Is very good. We get to resolutions very quickly. Our partner in Brazil is there for us every time.

I give it a nine out of 10. 3PAR is very good, but not perfect.

It serves as our centralized storage for three tiers of storage, and we have half a petabyte.

It performs very, very well.

We've had an overall improvement of performance in terms of actually doing transaction data, and we have a lot more capacity than we once had.

The most valuable feature for me is the support, three tiers of support. Storage is storage. The reality is, is it available and functional? Service and support are more important than capacity.

I would like to see compatibility with NVMe.

Resilience and reliability, unmatched. They take good care of us.

The scalability is far beyond what we'll ever need.

The service and support are far better than any other service and support I've had. They are very proactive and very sensitive to any kind of disruption of functionality.

Our data center was compromised and I had to replace the entire data center. Our previous solution was HPE LeftHand.

The most important criteria for me when selecting a vendor are

• cost
• reliability
• support.

The setup was complex from the perspective of the employees having to go through five days of training. If they simplified the administrative process of maintaining the unit, that would go a long way.

We have seen value when it comes to ROI but I can't elaborate.

Nimble. We went with 3PAR because we got a special deal on pricing.

I rate this solution an eight out of 10 because administrative functionality requires five days of training.

We use 3PAR as a shared platform for services, applications, and databases.

For the past three years, all-flash has been the most innovative way of using storage, based on the 3PAR platform.

It is scalable.

I am hoping to see some snapshot functionalities: Oracle snapshots, snapshots for VMware, and hypervisor snapshots.

We have had some bad issues on stability. Also, the new feature sets, like deduplication and compression, are complex to work with. I hope when I view the roadmap that they will be less complex.

We have a bad track record on the 3PAR system. We also have some Virtual Fibre Channel (VFC) storage platforms. However, we had a bad start on 3PAR. It is going better now.

We are managing more than 50 3PAR arrays worldwide. In our test centers, we are managing 15 HP all-flash arrays. The use of it is to be able to scale the platform. This is the most flexible way and one of the requirements for use in our environment.

Our administrator uses the support every day. So far, I hear that it is working okay. The Call-Hall function is what we are putting our value on. This is also the primary reason for using their support.

We were using EMC VMAX, EMC VNX, and NetApp. We switched due to price.

HPE has always done the initial hardware setup of the solution. It is straightforward, because we have good local engineers.

The performance compared to other products. Over time having the product and using it for six years, the performance has been a tremendous and bought about positive things.

We always channel all our vendors whether it is HPE or some others. Price is one of the most important features, but functionalities, availability, and the scalability of the systems is also very important.

Do some benchmarks on it and compare the requirements for whoever will be installing this product.

The number one feature is speed. We're serving almost 3.2 million people, we're running an eligibility system, we are on about 80 terabytes of data that need to be churned and processed.

In addition, the standard Oracle features like the compression. 

Right now, all of our development to pre-production, everything is in one single rack, which is easier for us to do. Production is on another rack, another full rack. That's what we use it for.

Pretty much whatever we need from the database side, it is there. 

There are specific things from the application side; I don't have a list, they do have a list of what you can see from the database side. 

Most of the developers are used to Microsoft .NET and SQL server. They are cutting out Oracle, so there is a shift within the developer's mind. "How am I going to, starting from the modelling, use it?"

Key items, like there's a 90-degree column in SQL that automatically fills in. It used to have triggers, but 12c is coming. There are some good features coming that I'm looking for. I was part of the session for what is new in 12c. I'm really excited about that. 

We're doing real-time analytics, so another thing is whether noSQL may be the best fit or not. We need to evaluate, not done yet. We are thinking, we have Exadata, we have Endeca, so we want to use the product.

Three and a half years, almost four years.

Absolutely stable. After we moved to Exadata the stability has been good. First, we used to go through Microsoft patching every month. With Exadata and Linux, we do only the firmware once in a quarter. It was really only required that we do it because we were behind two firewalls, so we didn't have to apply all the patches.

Absolutely scalable. We have our own five enterprise applications. They're all public-facing systems. Again, going back to the performance. On a given day, anywhere between 5000 to 8000 internal people use them. We do have a public-facing system. People apply for benefits, and it's entirely public, so they can use it. 

Compared to before, if you look at the performance, some of the applications are 30 to 40 percent improved. Some of the batches are almost 70% better. If you look at the back-end side, your export logical backups, and your disaster recovery, all those things, there is a tremendous increase.

Support is pretty good. There are certain areas specific to the enterprise application, we may not be getting all the support. But for Oracle products in general, pretty good.

Previously, we were on Sun Solaris, then we moved to a Windows platform, and it was really bad. The processing took too long. Start the load on Tuesday, Wednesday sometime around noon it would finish. Sometimes it would crash. Right now it is within four and a half hours. Starts at 6:00, 9:00 it's done. That is one big advantage we see.

We brought Oracle in to set it up. We didn't go through the entire setup process because Oracle was part of it. Oracle itself did it, so it was pretty easy for us. One good thing was that we didn't have to worry about the patching, OS levels, and other stuff. It was pretty easy for us.

When it comes to choosing a vendor, the primary criterion is support. When you have an issue, you need to immediately have a solution; it's a 24/7 application. For example, suppose we are going through patching, and something we don't anticipate comes up, I suddenly find an issue. I never saw that in my lower environments, and now production is impacted, so we create a severity-one ticket. How soon will it be resolved? can I trust it? That's number one. 

There may be some custom solutions we may be looking for that are not part of the product. Depending on what type of solution it is, we may need additional support. 

Also training. I think Oracle is growing fast. From '91 when I was with Oracle 6, I know how much it grew outside the databases. So training is one of the biggest issues that we are having. We moved from a Windows platform to a Linux platform, and also how to manage the entire thing. Enterprise Manager completely changed. So it is necessary to do some self-learning, but also vendor-provided training at a low-cost would help. Those are the things we are looking for. 

On top of that, there are certain things that would be of benefit. There is a new release coming in. If they can provide us with the documents, really help our DBA's to be on the cutting edge, rather than they themselves having to explore.

I didn't give it 10 out of 10 because there are still be a few issues that we are working on, but from a performance point of view, ease of use, scalability and reliability everybody is happy with the product.